Abstract: The elastic load on demand side can change with the fluctuation electricity price, so it could be modeled as the adjustable load cluster and involved into power dispatching with traditional generation unit, which can improve the flexibility of the power system. Due to the influence of various factors, such as the response to electricity price is not timely, signal transmission delay, the user's use of electricity habits is difficult to be changed, and the response amount and expected value always exist in a certain range of deviation. In this paper, the value at risk (VaR) is used to measure the risk loss caused by the uncertainty of the load cluster response, then VaR is introduced as a loss measure to participate in the optimization of the power dispatching model and establish a joint scheduling model. Through the actual example analysis, the results show that, due to the risk loss of the adjustable load cluster is considered, and the risk loss is reduced because of the uncertainty of the response. The research has practical application value to the implementation of DSM projects.